w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Fatigue of SpotFatigue of Spot--Welded Sheet Steel Joints:Welded Sheet Steel Joints:Physics, Mechanics, and Process VariabilityPhysics, Mechanics, and Process Variability
R. Mohan Iyengar1 , M. Amaya2, J. Bonnen3, K. Citrin2, H.-T. Kang4, S. Laxman1, A. Khosrovaneh5, N. Schillaci6, and H.-S. Shih7
1 Severstal North America Inc; 2Chrysler Corporation; 3 Ford Motor Company;4 University
of Michigan-Dearborn;5General Motors Corporation;6Arcelor-Mittal; 7 U. S. Steel
Great Designs in Steel – April 9, 2008
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Car Company Members:Mark Amaya Chrysler Corp.Ken Citrin Chrysler Corp.John Bonnen FordA.K. Khosravaneh General Motors
Steel Company Members:Rob Comstock AK SteelRaj Mohan Iyengar Severstal NATodd Link U.S. SteelH.-C. (Mike) Shih U.S. Steel Nick Schillaci Arcelor-MittalBenda Yan Arcelor-Mittal
Staff MembersBart Clark A/SP
Sheet Fatigue Team Members
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
• Weight reduction
What are the implications of replacing heavier gage components made of low carbon and HSLA materials with thinner AHSS parts?
• Over 5000 spot welds in a typical uni-body structure fastens panels and components together. Fatigue resistance of these joints impart significant influence on durability of a vehicle.
Motivation
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership Spot Weld Materials & Gages
Gagemm
IFGI
1.60
DQSKGI
1.59
CQSKBare1.83 1.74
HSLA
340GI
1.78
DP600GI
1.53
DP800GA1.61
DP980Bare1.55
TRIP
600Bare1.64
TRIP
800EG1.53
RA830GI1.39
MS1300Bare1.60
Boron
Bare1.74-1.47
156 1237
1382
5.55
5.55
350
14.5
22.7
YS, MPa 178 170 370 432 414 702 421 510 901 1156
UTSMPa 306 308 448 671 782 1057 672 839 895 1355
UE, % 21.7 20.5 15.9 13.6 12.7 7.1 20.6 22.3 0.8 3.2
TE, % 32.2 32.5 31.7 22.1 19.5 11.4 29.3 27.2 6.7 5.1
Grades & Gages
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Squeeze time: 99 cyclesWeld time: 22 cyclesHold time: 90 cyclesElectrode force: 1500 lb (DQSK, IF ~1100lb)Cap size, truncated: 7.9mm
Spot Welding Parameters for All Steels
Weld Current Adjusted for Each Material
Target Button Size: 7.0 mm
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Tensile ShearTensile Shear Coach Peel
Fatigue Specimen Designs
t
25mm
38mm
80mm
160mm
12.5mm
R = 2.5t
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
10
100
1000
10000
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to Failure
Load
Am
plitu
de (N
)
DQSK-R0.1DQSK-R0.3
CQSK-R0.1CQSK-R0.3
IF-R0.1IF-R0.3HSLA340-R0.1
HSLA340-R0.3DP600-R0.1
DP600-R0.3DP800-R0.1
DP800-R0.3DP980-R0.1DP980-R0.3
TRIP600-R0.1TRIP600-R0.3
TRIP800-R0.1TRIP800-R0.3RA830-R0.1
RA830-R0.3MS1300-R0.1
MS1300-R0.3Boron-R0.1
Boron-R0.3
Tensile Shear
Coach Peel
All Spot Weld Results (nominal -1.6mm)
Failure cycles
Loa
d,
N
Thickness effectNo mean stress effect
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
1mm
Untested Nugget
1.78mm HSLA-340Y-GI sheet.
Typical Fatigue fracture
250μm
Location of Crack Initiation and Growth
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
10
100
1000
10000
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to Failure
Load
Am
plitu
de (N
)
HSLA340TS01 1.78mmHSLA340TS03 1.78mmHSLA340CP01 1.78mmHSLA340CP03 1.78mmHSLA340TS01 1.00mmHSLA340TS03 1.00mmHSLA340CP01 1.00mmHSLA340CP03 1.00mmDP600TS01 1.53mmDP600TS03 1.53mmDP600CP01 1.53mmDP600CP03 1.53mmDP600TS01 0.83mmDP600TS03 0.83mmDP600CP01 0.83mmDP600CP03 0.83mm
Tensile Shear
Coach Peel
Thickness Effects
Failure cycles
Loa
d,
N
DP 600 – 0.83 mm
HSLA 340 – 1.78 mm
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
100
1000
10000
1.E+04 1.E+05 1.E+06 1.E+07 1.E+08Cycles to Failure
Max
. Loa
d (N
)
DP600 As-Received, 1.29mm thick, 7.0mm Dia. button
DP600 Prestrained, 1.19mm thick, 7.3mm Dia. button
R=0 loading
10
100
1000
1.E+04 1.E+05 1.E+06 1.E+07 1.E+08
Cycles to Failure
Max
. Str
ess
(MPa
)
DP600 As-Received, 1.29mm thick, 7.0mm Dia. buttonDP600 Prestrained, 1.19mm thick, 7.3mm Dia. Button
(2)
Normalized using Rupp parameter
Raw Fatigue Data
T.M. Link, 45th Mechanical Working and Steel Processing (MWSP) Conference Proceedings, Vol. XLI, 2003
DP600
DP600
Effects of Prestrain
Failure cycles
Loa
d,
NFailure cycles
Loa
d,
N
Minimal effect of pre-strainwhen thickness of formed sheet is accounted for
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
1.E+02
1.E+03
1.E+04
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to Failure
Load
Am
plitu
de (N
)
HSLA340TS01 SW
HSLA340TS03 SW
HSLA340CP01 SW
HSLA340CP03 SW
HSLA340TS01 Bond
HSLA340TS03 Bond
HSLA340CP01 Bond
HSLA340CP03 Bond
HSLA340TS01 WB
HSLA340TS03 WB
HSLA340CP01 WB
HSLA340CP03 WB
DP600TS01 SW
DP600TS03 SW
DP600CP01 SW
DP600CP03 SW
DP600TS01 Bond
DP600TS03 Bond
DP600CP01 Bond
DP600CP03 Bond
DP600TS01 WB
DP600TS03 WB
DP600CP01 WB
DP600CP03 WB
Tensile Shear
Coach Peel
SW = Spot WeldBond = Bond OnlyWB = Weld Bond
Spot Weld, Bond-Only, & Weld Bonded
Failure cycles
Loa
d,
N
blue – weld-bonded
red – bond only
black – spot weld
Coach peel
Tensile shear
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Tensile Shear Specimens, 1.78 mm HSLA 340-GI, R = 0.1
1000
10000
1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to Failure
Load
Am
plitu
de, N
7.0 mm Dia, 1 cycle HT7.0 mm Dia, 90 cycle HT7.0 mm Dia, 1 cycle HT, heat treated7.0 mm Dia, 90 cycle HT, heat treated7.0 mm Dia, 1 cycle HT, weld bonded7.0 mm Dia, 90 cycle HT, weld bonded
(3)
a)
Multiparameter Study:Hold Time, Heat Treatment, w/ Weld-bond
Tensile Shear Specimens, 1.78 mm HSLA 340-GI, R = 0.1
1000
10000
1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to Failure
Load
Am
plitu
de, N
4.9 mm Dia, 1 cycle HT4.9 mm Dia, 90 cycle HT4.9 mm Dia, 1 cycle HT, heat treated4.9 mm Dia, 90 cycle HT, heat treated4.9 mm Dia, 1 cycle HT, weld bonded4.9 mm Dia, 90 cycle HT, weld bonded
(4)
b)
HSLA 340 (1.78mm) HSLA 340 (1.78mm) –– Button Button DiaDia: 7 mm: 7 mm
HSLA 340 1.78 mm HSLA 340 1.78 mm –– Button Button DiaDia: 4.9 mm: 4.9 mm
Hold Time: Thought to be detrimental for AHSS
For HSLA 340:•Effect of HT or heat treatment – small for 7.0 mm & more (longer HT – better lives) for 4.9 mm welds•More scatter for 4.9 mm welds•Weld bond improves fatigue life
Weld Bond: Load transfer over a larger region
Heat Treatment: Paint & Bake Cycle
Failure cycles
Failure cyclesLoa
d,
N
Loa
d,
N
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Multiparameter Study:Hold Time, Heat Treatment, w/ Weld-bond
For DP 600:•Effect of HT or heat treatment – small for 7.0 mm & more (shorter HT – better lives) for 4.9 mm welds•More scatter for 4.9 mm welds•Weld bond improves fatigue life
Tensile Shear Specimens, 1.53 mm DP 600-GI, R = 0.1
1.E+03
1.E+04
1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to Failure
Load
Am
plitu
de, N
7.0 mm Dia, 1 cycle HT7.0 mm Dia, 90 cycle HT7.0 mm Dia, 1 cycle HT, heat treated7.0 mm Dia, 90 cycle HT, heat treated7.0 mm Dia, 1 cycle HT, weld bonded7.0 mm Dia, 90 cycle HT, weld bonded
(4)
a)
Tensile Shear Specimens, 1.53 mm DP 600-GI, R = 0.1
1.E+03
1.E+04
1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to FailureLo
ad A
mpl
itude
, N
4.9 mm Dia, 1 cycle HT4.9 mm Dia, 90 cycle HT4.9 mm Dia, 1 cycle HT, heat treated4.9 mm Dia, 90 cycle HT, heat treated4.9 mm Dia, 1 cycle HT, weld bonded4.9 mm Dia, 90 cycle HT, weld bonded
(3)
b)
DP600 (1.53 mm) DP600 (1.53 mm) –– Button Button DiaDia: 7 mm: 7 mm
DP600 (1.53 mm) DP600 (1.53 mm) –– Button Button DiaDia: 7 mm: 7 mm
Loa
d,
N
Loa
d,
NFailure cycles
Failure cycles
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Dong
Kang
Rupp
Evaluation of Fatigue DamageParameters
1.E+02
1.E+03
1.E+04
1.E+05
1.E+06
1.E+07
1.E+08
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08Experimental Fatigue Life (Nf)
Pred
icte
d Fa
tigue
Life
(Nf )
Tensile ShearCoach Peel
Rupp
x5
x5
1.E+02
1.E+03
1.E+04
1.E+05
1.E+06
1.E+07
1.E+08
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08Experimental Fatigue Life (Nf)
Pred
icte
d Fa
tigue
Life
(Nf )
Tensile ShearCoach Peel
Dong
x5
x5
1.E+02
1.E+03
1.E+04
1.E+05
1.E+06
1.E+07
1.E+08
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08Experimental Fatigue Life (Nf)
Pred
icte
d Fa
tigue
Life
(Nf )
Tensile ShearCoach Peel
Kang
x5
x5
Tensile Shear
Coach Peel
Rupp
Kang
Dong
Bonnen et al.,SAE Paper No. 2006-01-0978.
http://www.a-sp.org/publications.htm.
Uncertainty of Analysisor
Scatter of Data due to Geometric Variability of Spot Welds Joints?
Measured Life
Measured LifeMeasured Life
Pre
dic
ted
Lif
e
Pre
dic
ted
Lif
e
Pre
dic
ted
Lif
e
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Center to Grip Distance
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07Cycle to Failure
Max
imum
App
lied
Load
, N
DP600 - 1.49, TSDP800 - 1.59, TSDP600 - 1.53, CPRA830 - 1.39, CP
TS Specimens: R2 = 0.96
CP Specimens: R2 = 0.96
Fatigue Strength Variability
Minimal effect of center-to-grip distance variation for current data
t
25mm
38mm
80mm
160mm
12.5mm
R = 2.5t
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Sheet Thickness
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07Cycle to Failure
Max
imum
App
lied
Loa
d, N
DP600 - 0.93DP600-1.54
TS Specimens: R2 = 0.76
CP Specimens: R2 = 0.38
Effect of thickness variation for current data
Fatigue Strength Variability
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to Failure
Load
(N)/M
omen
t (N
mm
)
Load vs LifeMoment vs Life
R2 = 0.94
R2 = 0.98
DP 600 – 0.93 mm
Flange Length in CP specimens
Fatigue Strength Variability
t
25mm
38mm
80mm
160mm
12.5mm
R = 2.5t
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Crucial Geometric Parameters
Q
P
F
e
t/2
2r
M
z
x
t/2
e
α
( ) 2122 3 xzxM τσσ +=
2
6btM
x =σ
2
4DQ
xz πτ =
Von Mises Stress Shear Stress
Axial Stress
2
6btM
x =σ
DtP
xz =τ
Tensile Shear
Tensile Shear
Coach Peel
Coach Peel
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
1.E+02
1.E+03
1.E+04
1.E+05
1.E+06
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to failure, cycles
Max
imum
App
lied
Loa
d, N
DP600-0.93 DP600-1.54TRIP600-1.64 HSLA-1.78RA830-1.35 MS1300-1.60IF-1.60 DQSK-1.60DP800-1.60
R2 = 0.86
Maximum Load - Tensile Shear
Fatigue Strength Variability
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to failure, cycles
von
Mis
es S
tres
s, M
Pa
DP600-0.93 DP600-1.54TRIP600-1.64 HSLA-1.78RA830-1.35 MS1300-1.60IF-1.60 DQSK-1.60DP800-1.60
R2 = 0.89
Effective Stress -Tensile Shear
Fatigue Strength Variability
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Stress Intensity Factor Range -Tensile Shear
Fatigue Strength Variability
1.E+00
1.E+01
1.E+02
1.E+03
1.E+04
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to failure, cycles
Stre
ss In
tens
ity R
ange
( ΔK
I), N
/mm
^1.5
DP600-0.93 DP600-1.54TRIP600-1.64 HSLA-1.78RA830-1.35 MS1300-1.60IF-1.60 DQSK-1.60DP800-1.60
R2 = 0.90
Mohan Iyengar et al.,SAE Paper No. 2008-01-0698.
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to Failure
Max
imum
App
lied
Loa
d, N
DP600-0.93DP600-1.54TRIP600-1.64HSLA-1.78DQSK-1.60DP800-1.60
R2 = 0.49
Maximum Load - Coach Peel
Fatigue Strength Variability
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to Failure
von
Mis
es S
tres
s1, M
Pa
DP600-0.93DP600-1.54TRIP600-1.64HSLA-1.78DQSK-1.60DP800-1.60
R2 = 0.81
Effective Stress - Coach Peel
Fatigue Strength Variability
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
Cycles to failure, cycles
Stre
ss In
tens
ity R
ange
( ΔK
I), N
/mm
^1.5 DP600-0.93
DP600-1.54TRIP600-1.60HSLA-1.78DQSK-1.60DP800-1.60
R2 = 0.83
Stress Intensity Factor Range - Coach Peel
Fatigue Strength Variability
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Conclusions
1. Spot weld fatigue performance of studied steels (AHSS, HSLA, low carbon) appears to be insensitive to base metal composition, microstructure, and strength.
2. Spot weld fatigue behavior is mainly controlled by geometric factors such as sheet thickness and weld diameter.
3. Spot weld fatigue behavior is largely mean stress insensitive, for the mean stresses examined.
4. No effect of weld hold time (between 1 and a 90 cycles).
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
5. No effect of paint bake cycle.6. Adhesive bonding and weld bonding significantly
improve fatigue behavior over spot welding alone, although this improvement is in keeping with the actual increase in joint area gained by the addition of the adhesive layer.
7. Prestraining or stretch-forming the parent metal before spot welding has no impact on the fatigue performance of spot welded joints.
8. Principal spot weld damage parameters (Rupp, Dong, Swellam, Kang, & Sheppard) predict A/SP data equally well.
9. Crucial parameters controlling the mechanics and physics uncovered, thus reducing uncertainty in prediction.
Conclusions
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
More information?
A/S-P spot weld fatigue knowledge base (including detailed fatigue data, report,
microstructure, tensile data, etc.) freely available at:
http://www.a-sp.org
http://www.a-sp.org/database/custom/ASP%20Spot%20Weld%20Fatigue%20Project%202-7-06v1c.exe
w w w . a u t o s t e e l . o r g
Auto/SteelPartnership
Acknowledgements
Thanks to other members of the sheet steel fatigue team for work on spot welds:
• Todd Link – US Steel• Ron Soldaat - Arcelor-Mittal• Benda Yan - Arcelor-Mittal
Thanks to the Department of Energy for their support of this program.